Array-CGH identifies cyclin D1 and UBCH10 amplicons in ... - CiteSeerX

Endocrine-Related Cancer (2008) 15 801–815

Array-CGH identifies cyclin D1 and UBCH10 amplicons in anaplastic thyroid carcinoma Jia-Jing Lee1,2, Amy Y M Au 3, Theodoros Foukakis1, Michela Barbaro1, Nimrod Kiss1, Roderick Clifton-Bligh3 , Johan Staaf 4 , A˚ke Borg4, Leigh Delbridge3, Bruce G Robinson3, Go¨ran Wallin1, Anders Ho¨o¨g 2 and Catharina Larsson1 1 Department of Molecular Medicine and Surgery, Karolinska Institutet, Karolinska University Hospital, CMM L8:01, SE-17176 Stockholm, Sweden 2 Department of Oncology-Pathology, Karolinska Institutet, Karolinska University Hospital, SE-17176 Stockholm, Sweden 3 Kolling Institute of Medical Research, Royal North Shore Hospital, University of Sydney, Sydney, New South Wales, Australia 4 Department of Oncology, Lund University, Lund, Sweden

(Correspondence should be addressed to J-J Lee; Email: [email protected]; [email protected])

Abstract Anaplastic thyroid cancer (ATC) is a rare but highly aggressive disease with largely unexplained etiology and molecular pathogenesis. In this study, we analyzed genome-wide copy number changes, BRAF (V-raf sarcoma viral oncogene homolog B1) mutations, and p16 and cyclin D1 expressions in a panel of ATC primary tumors. Three ATCs harbored the common BRAF mutation V600E. Using arraycomparative genomic hybridisation (array-CGH), several distinct recurrent copy number alterations were revealed including gains in 16p11.2, 20q11.2, and 20q13.12. Subsequent fluorescence in situ hybridization revealed recurrent locus gain of UBCH10 in 20q13.12 and Cyclin D1 (CCND1) in 11q13. The detection of a homozygous loss encompassing the CDKN2A locus in 9p21.3 motivated the examination of p16 protein expression, which was undetectable in 24/27 ATCs (89%). Based on the frequent gain in 11q13 (41%; nZ11), the role of CCND1 was further investigated. Expression of cyclin D1 protein was observed at varying levels in 18/27 ATCs (67%). The effect of CCND1 on thyroid cell proliferation was assessed in vitro in ATC cells by means of siRNA and in thyroid cells after CCND1 transfection. In summary, the recurrent chromosomal copy number changes and molecular alterations identified in this study may provide an insight into the pathogenesis and development of ATC. Endocrine-Related Cancer (2008) 15 801–815

Introduction Anaplastic thyroid cancer (ATC) is one of the most aggressive human malignancies, with a median survival of 3–6 months after diagnosis (Kondo et al. 2006). It is relatively rare comprising up to 5% of all thyroid cancers and mainly affects the elderly (Kondo et al. 2006). The natural history of the disease is characterized by rapid and uncontrolled local growth eventually causing suffocation. Distant metastases frequently develop and are mainly located in the lung. The treatment commonly involves radiotherapy and chemotherapy that are given pre-operatively followed by a surgery (Wallin et al. 2004). Little is presently known about the cellular origin and molecular etiology of ATC. This is partly attributed to the extensive necrosis that is characteristic of the disease

and further augmented by the pre-operative treatment. In some patients, a differentiated thyroid cancer is found adjacent to the ATC. Furthermore, in few cases, there is a continuous spectrum from differentiated to poorly differentiated thyroid cancer (PDTC) and ATC in support of a progression model. TP53 mutations that are rare in well-differentiated thyroid cancers, i.e., papillary thyroid cancer (PTC) and follicular thyroid cancer (FTC), are frequent in PDTC and reach up to 68% in ATC (Kondo et al. 2006). PIK3CA mutations have been reported in 12–23% of ATC (Garcia-Rosta´n et al. 2005, Hou et al. 2007). Activating mutations of RAS (that are mainly seen in FTC) and BRAF (V-raf sarcoma viral oncogene homolog B1), which characterizes aggressive PTC, are also found in a subset of ATCs while RET (rearranged during transfection)/PTC and PAX8

Endocrine-Related Cancer (2008) 15 801–815 1351–0088/08/015–801 q 2008 Society for Endocrinology Printed in Great Britain

DOI: 10.1677/ERC-08-0018 Online version via http://www.endocrinology-journals.org

J-J Lee et al.: Genetic alterations in anaplastic thyroid cancer (paired box gene 8)/PPARg (peroxisome proliferatoractivated receptor-g) rearrangements have not been determined in ATC (Kondo et al. 2006). Studies of gene copy number imbalances in ATC using conventional CGH have demonstrated recurrent gains of chromosomal regions 3, 5p, 11q13, and 20q and losses at 5q11–31 and Xp (Wreesmann et al. 2002, Rodrigues et al. 2004). Recently, we characterized karyotypic abnormalities and copy number alterations in ATC cell lines, which revealed gain of 20q as the most common abnormality (Lee et al. 2007). Here, we used bacterial artificial chromosome (BAC) arrays with whole-genome tiling resolution to investigate the DNA copy number alterations in a series of primary ATCs. Subsequently, we investigated the involvement of candidate genes located in areas of recurrent changes.

Materials and methods Established cell lines

The human ATC lines (HTh 104, HTh 112, HTh 7, HTh 74, C 643, KAT-4, SW 1736, ARO, and HTh 83) and Nthy-ori 3-1 (SV-40 immortalized normal human thyroid follicular cells; ATCC, Manassas, VA, USA) were cultured under conditions as described previously (Lee et al. 2007). Patients and tumor tissues

Fresh-frozen primary tumors from 28 cases of ATC (Table 1) were collected at the Karolinska University Hospital, Stockholm, Sweden and the Royal North Shore Hospital, Sydney, Australia. Twenty-three patients had received pre-operative radiotherapy and/or chemotherapy according to standard treatment protocols

Table 1 Clinical information for the 28 cases of primary anaplastic thyroid cancer (ATC) in the study Follow-up

Case no.

Sex M/F

Age at diagnosis (yrs)

Time

Preop. Other thyroid therapy tumors

Tumor size (cm)

Metastases/local invasion at diagnosis

Outcome

1 2 3 4 5 6 7 8

F F F F F M M M

75 85 77 83 62 82 68 72

D D D D D D D D

5.5 m 4.0 m 3.0 m 11.0 m 5.5 m 1.0 m 3.0 m 1.5 m

No Yes Yes Yes Yes Yes Yes Yes

No FTC (cs) PTC No No PTC No No

8.5 4.0 7.0 6.0 6.0 4.0 5.0 10.0

62 72 84 78 51 84 72 80 77 70 54 81 91 68

D D D A D D D D D D D D D D

0.3 m 3.5 m 3.0 m 66.0 m 1.0 m 3.0 m 1.0 m 0.5 m 0.5 m 1.0 m 11.0 m 4.0 m 2.5 m 3.0 m

Yes Yes Yes Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes

No FTA No No No No No PTC No FTA No No PDTC (cs) PTC

4.5 6.5 5.0 8.5 7.0 10.0 10.0 8.0 8.0 5.0 13.0 5.0 5.0 3.5

81 83 73 52 72 72

D D D D NA D

1.0 m 9.0 m 3.5 m 6.0 m NA 1.0 m

Yes Yes Yes No NA NA

No No No No PTC No

5.0 3.5 6.0 6.0 6.5 7.0

No No No No No NA Lung, lymph node Lung, lymph node, local invasion Lung Lung No No No No No No Lung No Local invasion No Lung, liver Lung, left adrenal, local invasion Lymph node No Lung, local invasion Local invasion Local invasion Lung

9 10 11 12 13 14 15 16 17 18 19 20 21 22

F M F F F M F F M F M F F M

23 24 25 26 27 28

F F M F M F

Previous goiter No Yes Yes Yes No No Yes No No Yes No No No Yes Yes No Yes Yes No No No No No No No No Yes Yes

F, female; M, male; yrs, years; m, months; NA, not available; D, dead; A, alive; Preop., preoperative; FTC, follicular thyroid cancer; PTC, papillary thyroid cancer; FTA, follicular thyroid adenoma; PDTC, poorly differentiated thyroid cancer; cs, continuous spectrum.

802

www.endocrinology-journals.org

Endocrine-Related Cancer (2008) 15 801–815 (Wallin et al. 2004). The histopathological diagnosis was established according to WHO classification (DeLellis et al. 2004), including findings of undifferentiated cells, giant and/or spindle cells, mitosis, and signs of necrosis. Tissue sampling and representativity testing followed established routines for the endocrine biobank. Frozen samples of medullary thyroid cancer and parathyroid adenoma were similarly collected at Karolinska University Hospital, Stockholm, Sweden, and used as references in western blot analyses. Informed consents were obtained from all patients and ethical approvals were granted. DNA and protein extractions

Tumor DNA was isolated by conventional methodology including phenol purification and ethanol precipitation. Cell line DNA was isolated as described previously (Lee et al. 2007). DNA was quantified using NanoDrop ND1000 (NanoDrop Technologies, Wilmington, NC, USA). BRAF mutation screening

The mutation hot spot exons 11 and 15 of BRAF were sequenced on both strands in all 28 tumors. The experimental procedure, amplification conditions (35 cycles), primers, and positive control were as previously described (Lee et al. 2007). Array-CGH analysis

Generation, hybridization, and analyses of the 33 K microarrays (resolution of 100 kb) with complete genome coverage produced by the SCIBLU Genomics, Department of Oncology, Lund University, Sweden (http://www.lth.se/sciblu) were essentially as previously reported for 32 K arrays (Barbaro et al. 2007). Genomic DNA of the tumor and commercial reference samples (Promega Corporation, Madison, WI, USA) was labeled as described previously (Jo¨nsson et al. 2007). Arrays were scanned using Axon GenePix 4200A microarray scanner (Molecular Devices, Sunnyvale, CA, USA). Individual spots identified on scanned arrays were collected using GenePix Pro 6.0 (Axon Instruments, Foster City, CA, USA), and the quantified data were loaded into Bio Array Software Environment (BASE; Saal et al. 2002). A BASE implementation of CGH-Plotter was used to identify regions of gains and losses after smoothing with a sliding window over three clones (Autio et al. 2003). Cut-off ratios for gains and losses constant were set at Z1.15 and K0.87 respectively, corresponding to log2 (ratio) of G0.2. A log2 (ratio) below K0.75 was www.endocrinology-journals.org

considered as homozygous loss and a ratio above C0.75 as amplification. Fluorescence in situ hybridization (FISH)

FISH analyses were performed on interphase imprints from frozen ATCs and on metaphase preparations of ATC cells using the BAC clone RP11-344G20 covering UBCH10 at 20q13.12 plus a chromosome 20 centromere probe (CEP20) as described previously (Lee et al. 2007), or pre-labeled probes for CCND1 and the chromosome 11 centromere (LSI CCND1 Spectrum Orange/CEP 11 SpectrumGreen1 Vysis, Inc., Downers Grove, IL, USA). Locus gain in a tumor was considered when a higher number of signals were recurrently observed for the gene-specific probe when compared with the centromere probe. Results for UBCH10 analysis of ATC cell lines have been published in Lee et al. (2007). Multiplex ligation-dependent probe amplification (MLPA) analysis

Three regions with prominent gains detected from the array, 20q11.2, 20q13.12, and 16p11.2, were selected for verification by MLPA. MLPA reactions were performed as described (Barbaro et al. 2007) using newly designed 5 0 and 3 0 half-probes targeting unique exonic or intronic sequences of genes within 20q11.2–q13.2 and 16p11.2 and control genes ALB (4q13.3) and CLDN16 (3q28) according to Barbaro et al. (2007) (Supplementary Table 1, which can be viewed online at http://erc.endocrinology-journals.org/ supplemental/). For each sample, the peak areas corresponding to each probe were first normalized to the average of the peak areas of the control probes, and then normalized to the average peak area in eight controls (normal lymphocyte DNA). Western blot analysis

Total protein extracts from tumor tissues (75 mg) and cultured cell protein were electrophoresed and transferred to nitrocellulose filters (Invitrogen, Carlsbad, CA, USA). For transfected cells, an aliquot was taken from the cell suspension where the starting cell count was 1!106 cells/well plated in a six-well plate. The filters were stained with Ponceau Red (Sigma) as a control for protein presence and incubated overnight at 4 8C with anti-cyclin D1 (1:400; SP4 clones; NeoMarkers, Fremont, CA, USA), anti-p16 (1:100; G175-405; BD PharMingen, San Jose, CA, USA), and anti-aactinin (1:100; AT6/172 clone; Chemicon International, Temecula, CA, USA) or anti-a-tubulin 803

J-J Lee et al.: Genetic alterations in anaplastic thyroid cancer (1:2500; Clone DM 1A; Sigma–Aldrich). Anti-aactinin and anti-a-tubulin served as loading controls. Cell proliferation analysis of cells overexpressing CCND1

Amaxa nucleofection technology (Amaxa Biosystems, Cologne, Germany) was used to transfect cells with siRNA and plasmids for MTS assays. For cell proliferation assays, 1!104 cells/well were plated (96-well plate) and, for western blot analyses, 1!106 cells/well were plated (6-well plate). HTh 7 cells were transfected with 1.5 mg siRNA/1!106 cells using program X-001 with the V solution and Nthy-ori 3-1 cells were transfected with 2 mg plasmid/1!106 cells using program A-020 with the T solution. Cells were incubated for 16 h prior to subsequent analyses at 0, 24, 48, and 72 h after overnight transfection. CCND1 siRNA (#SI02654540, Qiagen GMbH, Valencia, CA, USA) was used in knockdown studies with the All Stars siRNA (#SI1027281, Qiagen) as negative control. The CCND1 plasmid was obtained from Addgene (Rc/CMV-CCND1 #8962, Cambridge, MA, USA) and the control plasmid Rc/CMV was kindly provided by Dr Sue Firth at the Kolling Institute of Medical Research, NSW, Australia. Successful transfection and siRNA were verified by western blot analysis and quantitative real-time PCR (qRT-PCR) as previously described (Lee et al. 2007). TaqMan Gene Expression Assays (Applied Biosystems, Foster City, CA, USA) were used to quantitate CCND1 (Hs00277039-m1) and 18S (#4319413E). Cyclin D1 expression from western blot analyses was quantified against a-tubulin expression by Multi gauge V3.0 (FujiFilm Global, Valhalla, NY, USA). Cell proliferation was quantitated by MTS assay (Promega Corporation) as per protocol at 0, 24, 48, and 72 h after overnight transfection. Absorbance readings (OD490–650) were taken 2 h after the addition of the MTS reagent. Statistical analysis

Potential correlations between the most frequently altered regions (11q13, 20q11.2, 20q13.12, 13q21.2–q21.31, and 16p11.2) and CCND1 copy number detected by FISH, cyclin D1 and p16 expressions, BRAF mutations, and association with PTC were investigated using Fisher’s exact test (n!5) or two-tailed c2-test (nO5), (http://www.graphpad.com/ quickcalcs/contingency1.cfm). For the validation of the 20q11.2, 20q13.12, and 16p11.2 regions identified from the array by MLPA, f-correlation was computed (Statistica version 6; StatSoft Inc., Johannesburg, 804

South Africa). Absorbance values of CCND1 siRNA knockdown or CCND1-transfected cells were compared with reference-treated cells using paired t-test. P values below 0.05 were considered significant.

Results Clinical characteristics and BRAF mutations in primary ATCs

The clinical characteristics of the 28 cases of primary ATCs studied are given in Table 1, and the molecular analyses carried out in individual cases are detailed in Table 2. Nine cases presented an additional thyroid cancer that was either adjacent or growing in continuous spectrum with the ATC. Three of the twenty-eight ATCs exhibited a heterozygous nucleotide alteration GTG/GAG at position 1799 in exon 15 of BRAF that leads to a missense mutation V600E (Table 2). Two of these cases had an additional thyroid cancer; case 6 exhibited a PTC adjacent to the ATC and, in case 21, a continuous tumor spectrum from PDTC to ATC was observed. Array-CGH analysis of primary ATCs

DNA copy number changes were detected in all 27 primary ATCs successfully studied by array-CGH, preferentially involving sub-chromosomal regions and gains (Tables 2 and 3; Fig. 1). Alterations that were commonly observed and further examined in this study include gains at 11q13, 20q, and 16p11.2 (Figs 2, 3 and 4A). Other frequent events observed (O20% of cases) include gains at 6p, 7q, 12q, 17q, 19, and 22q, and losses on 4q and 13q. Gains in telomeres were observed in O20% of the ATC panel for most chromosomes with the exception of chromosomes 2, 3, 6, and 15. The smallest regions of overlap (SRO) identified by alignment of all alterations in individual tumors are summarized in Table 3. Close to half of the panel showed gain in 11q13 and the associated SRO at 11q12.2–q13.2 was altered in 33% of the tumors (Table 3; Fig. 2A). Gain in 20q was one of the most frequently observed aberrations (nZ14; Table 2), where 52% of the panel showed gain in either 20q11.2 or 20q13.12 (Fig. 3A). Gain of 16p11.2 was observed in 48% of the tumors (Fig. 4A; Table 3). In the long arm of chromosome 22, gains were seen in the two separate regions 22q11.21 (nZ16) and 22q13.1 (nZ15; Table 3). Similarly gain of chromosome 19 involved the 19p13 region (67% of the tumors) as well as 19q13.1–q13.2 (52%). Gain in 7q11.22–q11.23 was also recurrently observed in the panel (52%; Table 3). www.endocrinology-journals.org

www.endocrinology-journals.org

Table 2 Molecular and genetic alterations of primary anaplastic thyroid cancer (ATC) detected in this study

BRAF sequence Ex 11C15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

wt wt wt wt wt V600E/wt wt wt wt wt wt wt wt wt V600E/wt wt wt wt wt wt V600E/wt wt wt wt wt wt wt wt

No FTC (cs) PTC No No PTC No No No FTA No No No No No PTC No FTA No No PDTC (cs) PTC No No No No PTC No

Expression by western blot

UBCH10

Cyclin D1

Yes K Yes Yes K Yes K Yes Yes K K K ND K K K K K K K ND K K Yes ND Yes ND Yes

CCND1

K K K Yes Yes K Yes K Yes K K K ND Yes K K K K K K Yes Yes Yes Yes ND K Yes K

C C CC CCC C CCC (K) CCC CCC (K) (K) CCC (K) (K) CC CCC CCC C C CC (K) (K) CC CC (K) ND C (K)

p16

(K) (K) (K) (K) (K) (K) (K) (K) (K) Yes (K) (K) (K) (K) Yes Yes (K) (K) (K) (K) (K) (K) (K) (K) (K) ND (K) Yes

Gain by array-CGH within chromosome arm (sub-band)

Array-CGH losses in

1q

6p

7q

11q

12q

16p

17q

19p

19q

20q

20q

22q

21

22–21

11.22–11.23

13

13

11.2

21

13

13.1–13.2

11.2

13.12

11.21

13.1 12–13.1 28.3

21.2–21.31

K K K K Yes K Yes K K K K K ND Yes K K K Yes K K K K K Yes K K Yes K

K K K Yes Yes K Yes K K K K K ND Yes K Yes K Yes Yes K Yes Yes Yes Yes Yes K Yes Yes

K K K Yes Yes K Yes K Yes K K K ND Yes K K K Yes K K Yes K Yes Yes Yes K Yes K

K Yes K Yes Yes K K Yes K K K Yes ND K K Yes K Yes Yes K Yes Yes Yes K Yes Yes K K

K K Yes K Yes K Yes K K K K K ND Yes K Yes K Yes Yes Yes Yes Yes Yes Yes K Yes K K

K K K Yes Yes K K K K K Yes Yes ND Yes K K K Yes Yes Yes Yes Yes Yes Yes Yes K K K

K K Yes Yes Yes K Yes K K K K Yes ND Yes Yes Yes K Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes K

K Yes Yes K Yes Yes K K K Yes K K ND Yes K K K Yes K Yes Yes Yes Yes Yes K K Yes Yes

K K Yes K Yes K K K Yes Yes K K ND K K K K Yes K K Yes Yes Yes Yes Yes Yes Yes K

K K Yes Yes Yes K K K Yes K K K ND Yes K K K K K K K K K Yes Yes K Yes K

K Yes K K Yes K K Yes Yes Yes K K ND Yes K Yes K Yes Yes K Yes Yes Yes Yes Yes K Yes Yes

K Yes K K Yes K Yes Yes K K K K ND Yes K Yes K Yes K Yes Yes Yes Yes Yes Yes K Yes Yes

K K Yes K K Yes K K K K K K ND K K Yes K K Yes K K K K K K K Yes Yes

K K K Yes Yes K Yes K K K K K ND K K K K K K K K K Yes K Yes K K K

22q

4q

K K K K K K K K K Yes K K ND K K K K Yes Yes K K Yes Yes K Yes K Yes Yes

4q

K K K K K Yes K K Yes Yes K K ND K K Yes K Yes K K K Yes Yes K Yes K Yes K

13q

‘K’, Not present; Yes, present; ND, not determined; ‘(K)’, not detectable; ‘C’, low; ‘CC’, moderate; ‘CCC’, high; FTC, follicular thyroid cancer; PTC, papillary thyroid cancer; FTA, follicular thyroid adenoma; PDTC, poorly differentiated thyroid cancer; cs, continuous spectrum. Locus gain by FISH: UBCH10OCEP20 or CCND1OCEP11.

805

Endocrine-Related Cancer (2008) 15 801–815

Other Case thyroid tumors no.

Locus gain by FISH

Frequent alteration

Cytoband Copy number gains C1q21 C6p22–p21 C7q11.22–q11.23 C11q13 C12q13 C16p11.2 C17q21 C19p13 C19q13.1–q13.2 C20q11.2 C20q13.12 C22q11.21 C22q13.1 Copy number losses K4q12–q13.1 K4q28.3 K13q21.2–q21.31

SROs No. of cases (Frequency)

Cytoband

5 (19%) 6 (22%) 14 (52%) 11 (41%) 13 (48%) 13 (48%) 13 (48%) 18 (67%) 14 (52%) 12 (44%) 8 (30%) 16 (59%) 15 (56%)

1q21.1–q21.3 6p21.32 7q11.22–q11.23 11q12.2–q13.2 12q13.11–q13.12 16p11.2 17q21.33 19p13.2 19q13.12 20q11.21–q11.22 20q13.12 22q11.21 22q13.1

14 65 14 610 3 31 65 527 7 20 35 816 6 33 21 142 4 73 84 342 2 80 82 639 4 59 49 240 1 19 50 033 4 05 42 097 3 14 97 814 4 18 87 694 1 95 38 297 3 60 03 989

(RP11-437M17) (RP11-521B19) (RP11-91L7) (RP11-289J6) (RP11-579D7) (RP11-410P5) (RP11-121F10) (RP11-566N12) (RP11-166B11) (RP11-120F10) (RP11-809G24) (RP11-54C2) (RP11-7I9)

14 81 02 224 3 34 94 594 7 22 59 198 6 62 12 707 4 86 39 263 2 89 22 556 4 64 66 966 1 21 50 103 4 12 11 643 3 31 84 582 4 41 53 298 1 99 39 137 3 64 44 264

(RP11-787J14) (RP13-512P23) (RP11-667P12) (RP11-775M2) (RP11-160B8) (RP11-674B7) (RP11-21I9) (RP11-754E16) (RP11-532B13) (RP11-612A10) (RP11-124K8) (RP11-818K20) (RP11-569A18)

1.59 0.33 0.22 2.89 1.25 0.84 0.52 0.20 0.63 1.69 2.27 0.40 0.44

3 (11%) 3 (11%) 13 (48%) 9 (33%) 8 (30%) 12 (44%) 8 (30%) 15 (56%) 13 (48%) 5 (19%) 8 (30%) 12 (44%) 13 (48%)

8 (30%) 9 (33%) 6 (22%)

4q12–q13.1 4q28.3 13q21.2–q21.31

6 13 17 501 13 20 76 562 6 11 13 549

(RP11-687A20) (CTD-2390L24) (RP11-418D23)

6 60 95 641 13 31 23 563 6 34 81 793

(RP11-257M8) (CTD-200F17) (RP13-495B8)

4.78 1.05 2.37

8 (30%) 6 (22%) 6 (22%)

bp start position (BAC)

bp end position (BAC)

Size (Mb)

No. of cases (Frequency)

J-J Lee et al.: Genetic alterations in anaplastic thyroid cancer

806 Table 3 Frequently altered regions and associated smallest overlapping regions (SROs) in primary anaplastic thyroid cancers (ATCs)

www.endocrinology-journals.org

Endocrine-Related Cancer (2008) 15 801–815

Figure 1 Frequency plot showing sequence copy number alterations detected in 27 primary ATCs for chromosomes 1–22. Gains are depicted as green and losses as red vertical bars representing one ATC case each. Candidate genes and regions selected for further analyses and the methods used are indicated below the plot.

Results from MLPA analysis (Supplementary Fig. 1, which can be viewed online at http://erc.endocrinologyjournals.org/supplemental/) were concordant with gain observed by array-CGH for 20q11.2 (fZ0.55), 20q13.12 (fZ0.51), and 16p11.2 (fZ0.54). Copy number losses were most frequently observed in 4q12–q13.1 (30%), 4q28.3 (33%), and 13q21.2–q21.31 (22%; Table 3). Other recurrent alterations are gains at 1q21, 6p22–q21, 12q13, and 17q21 (Table 3). Interestingly, ATC with concomitant PTC frequently showed loss at 13q21.2–q21.31 (PZ0.003; Fisher’s test). Comparison between genetic alterations with clinical parameters (sex, age at diagnosis, survival, preoperative treatments, tumor size, metastasis, and goiter) did not reveal additional statistically significant correlations.

Amplifications and prominent losses revealed by array-CGH in primary ATCs

Interestingly, high-level amplifications with log2 (ratio) exceeding C0.75 were identified in chromosomes 11, 18, and 20 for cases 7, 3, and 4 respectively. ATC case 7 exhibited high-level amplification of a 5 Mb region in 11q22.1 (log 2 (ratio)Z1.5–2.0; Fig. 2A). Similarly, a 6 Mb region in 18q11.2 was highly amplified in ATC case 3 (log2 (ratio)Z1.2; Fig. 4B). Finally, in ATC case 4, high-level amplification was observed of the commonly altered 20q13.12 region that includes the UBCH10 candidate gene (Fig. 3A). A homozygous loss of a 5 Mb region was detected in ATC case 6 encompassing the CDKN2A gene locus in 9p21.3 (Fig. 4D). Furthermore, almost identical small regional deletions within 5q were observed in cases 1 and 11, which included a common 1.9 Mb region in 5q13.2 (Fig. 4C). In these two cases, other regions www.endocrinology-journals.org

commonly altered in the ATC panel were largely unaffected (Table 2). Frequent lack of p16 expression in ATCs

Since a homozygous loss at the 9p21.3 locus encompassing the CDKN2A gene was detected in case 6, it was of interest to confirm the presence of tumor suppressor p16 encoded by this locus for case 6 as well as to further investigate p16 expression in the entire panel by western blot analysis. The p16 protein expression was detected in positive control cells SAOS-2 osteo sarcoma and in medullary thyroid carcinoma tissue, but was not detectable in normal thyroid or MCF-7 cells (negative control; Fig. 4D). In addition, no p16 expression was observed in 24 out of 27 ATCs analyzed (89%; Table 2), including case 6 with homozygous loss at the CDKN2A locus. Gains of chromosome 20 and locus gain of UBCH10

Two separate regions of copy number gain were observed for chromosome 20, of which the more distal at 20q13.12 encompasses the UBCH10 gene that has been suggested to be associated with ATC (Pallante et al. 2005, Lee et al. 2007). We therefore performed a dual-color FISH analysis on ATC imprints using a CEP20 and a BAC clone containing UBCH10. Out of 24 samples, 9 (38%) showed an increased copy number for UBCH10 (Table 2; Fig. 3B). Gain of CCND1 in 11q13 and overexpression of cyclin D1 protein

Amplification of chromosomal region 11q13 is associated with gain for CCND1 in several human cancers (Alao 2007). To determine whether the CCND1 gene is gained in ATC tumors and cell lines, 807

J-J Lee et al.: Genetic alterations in anaplastic thyroid cancer

Figure 2 (A) Array-CGH profiles of chromosome 11 for case 7 (upper) carrying amplification in 11q22.1 and gain in 11q13, and for case 4 (below) harboring an 11q13 amplicon. (B) Western blot analyses showing cyclin D1 expression in positive control cells (C), and ATCs 21, 25, and 27, while normal thyroid (N) and ATCs 22, 23, and 24 are negative. Incubation of the same filter with a-actinin served as loading control. (C) Fluorescence in situ hybridization (FISH) of CCND1 (Cyclin D1, red) and centromere 11 (CEP11, green) copy numbers. Two signals are observed in normal metaphase and interphase nuclei, while ATC cases 4 and 7, and HTh 7 cells show relative gain of CCND1.

dual-color FISH analysis was performed with a CCND1 clone and centromere 11 (CEP 11) as a control for chromosome copy number. As illustrated in Fig. 2C, locus gain of CCND1 when compared with CEP 11 was recurrently observed in interphase nuclei of 38% of the ATCs (Table 2). The observation of CCND1 locus gain coincides with the presence of 11q13 gain by array-CGH (PZ0.0001; two-tailed c2-test). Locus gain of CCND1 was also recurrently observed in interphase and metaphase cells of the ATC line HTh 7 (Fig. 2C). Western blot analysis showing strong cyclin D1 expression in parathyroid tumor tissue was used as a positive control, while in normal thyroid cyclin D1 expression was not detectable (Fig. 2B). In primary 808

ATCs, varying levels of cyclin D1 expression were observed in 18 out of 27 cases studied (Table 2). Among these 18 ATCs, the protein level was determined as low (six cases), intermediate (five cases), or high (seven cases) as exemplified in Fig. 2B. Effects of CCND1 on proliferation of thyroid cells

Gain of CCND1 or cyclin D1 overexpression was observed in the majority of ATCs, while its possible influence on thyroid cell proliferation was assessed in ATC cells (HTh 7) and normal human thyroid cells (Nthy-ori 3-1). The HTh 7 cells showed regional gain of the CCND1 locus and overexpression of cyclin D1 protein and were therefore selected for transfection with small interfering RNA (siRNA) oligonucleotides www.endocrinology-journals.org

Endocrine-Related Cancer (2008) 15 801–815

Figure 3 (A) Array-CGH profiles of chromosome 20 for case 5 (upper), showing gain in 20q11.2 and 20q13.12, and case 4 (below) with high-level amplification in 20q13.12. (B) FISH analysis of UBCH10 (green) and centromere 20 (CEP20, red) in normal metaphase and interphase nuclei, and in ATC cases 8, 9, and 3 with regional gain of UBCH10.

www.endocrinology-journals.org

809

J-J Lee et al.: Genetic alterations in anaplastic thyroid cancer

Figure 4 (A) Array-CGH profile exemplifying the frequent gain of chromosome 16 in case 7, harboring a gain at 16p11.2. (B) ArrayCGH profile of chromosome 18 for case 3 with amplification at the 18q11.2 region. (C) Array-CGH profile of chromosome 5, highlighting narrow deletion in 5q13.2 in case 1 (upper) and case 11 (below). (D) Array-CGH profile of chromosome 9 for case 6, harboring a homozygous loss at 9p21, where the CDKN2A gene is located. Western blot analyses show p16 expression in medullary thyroid cancer (MTC), ATC cases 10 and 28, and SAOS-2 cells used as a positive control (C). The p16 expression is not detected in MCF-7 cells (K), normal thyroid (N) or ATC cases 12, 11, 5, and 6. Subsequent incubation of the same filter with a-actinin served as loading control.

against CCND1. Successful siRNA within 24 h was demonstrated by 20–30% decrease in western blot expression and up to 60% decrease in CCND1 mRNA expression by qRT-PCR (Fig. 5A). Slightly lower proliferation measured by MTS assay absorbance was observed after CCND1 siRNA when compared with All Stars siRNA used as a control (Fig. 5). Transfection of Nthy-ori 3-1 cells with a cyclin D1 expressing construct resulted in stable 3-fold increase in protein expression and 12-fold increase in mRNA levels (Fig. 5B). Only minor increase in proliferation was observed in cyclin D1 expressing cells when compared 810

with control cells transfected with empty vector (Fig. 5). Taken together, CCND1 siRNA and transfection assays had only minor effects on proliferation, which were not statistically significant.

Discussion This is the first report of genome-wide detection of DNA copy number changes in ATC using array-CGH. A multitude of recurrent changes were detected in the 27 ATCs, and the role of candidate genes in selected www.endocrinology-journals.org

Endocrine-Related Cancer (2008) 15 801–815

Figure 5 Analysis of CCND1 (cyclin D1) effect on thyroid cell growth. Comparison plot for proliferation (MTS) assay (average of three independent experiments) of (A) HTh 7 ATC cells transfected against CCND1 siRNA and its reference control cells at 0, 24, 48, and 72 h after transfection, and (B) Nthy-ori 3-1 normal thyroid cells transfected with CCND1 or empty vector (pRC/CMV). Standard errors are indicated at each time point. Efficiency of siRNA as well as CCND1 transfection was validated by western blot analysis and qRT-PCR. Quantification was performed against a-tubulin for the western blot analysis and against 18S for qRT-PCR.

areas of chromosomal gain (CCND1 and UBCH10) or loss (p16) was further examined. An average of 44 DNA copy number changes was detected in each tumor, which is considerably higher than the changes found in differentiated thyroid cancer (Hemmer et al. 1999, Kjellman et al. 2001, Wreesmann et al. 2002, 2004, Rodrigues et al. 2004). This was rather expected, as aggressive and advanced cancers are generally genetically unstable. The widespread telomeric gains observed in this study is an uncommon property of differentiated thyroid tumors, www.endocrinology-journals.org

further supporting chromosomal instability in ATC. A role for telomere dysfunction in promoting gene amplification and hence chromosome instability, which is the hallmark of human cancer, is supported by tumor-bearing mice model (Albertson 2006). It could be argued that the observed copy number alterations are a result of the pre-operative treatment administered to the patients. However, the number and patterns of changes in the three patients, which were operated primarily without pre-treatment did not differ from the rest. Furthermore, many of the alterations 811

J-J Lee et al.: Genetic alterations in anaplastic thyroid cancer were detected recurrently across the different tumors, while alterations resulting from pre-operative radiochemotherapy are expected to be more random. Gain of 11q13 was found by array-CGH and FISH analysis for the CCND1 locus in w50% of ATCs. This finding is in agreement with previous reports of 11q13 gain in ATC lines (Lee et al. 2007) and ATC primary tumors (Wreesmann et al. 2002), and motivated further investigation of the known oncogene CCND1 encoding cyclin D1. In protein studies, 67% of ATCs were shown to express cyclin D1 while no expression was observed in normal thyroid tissue. Cyclin D1 expression has been reported to be especially prevalent in aggressive forms of thyroid cancers (Wang et al. 2000, Khoo et al. 2002). Gain of 11q13 and/or CCND1 is also frequent in other tumor types of relatively advanced stage, including breast, head, and neck as well as esophagus carcinomas, sometimes as part of homogenously staining regions (Arnold & Papanikolaou 2005). Notably, cyclin D1 overexpression in this study occurred both in presence or absence of 11q13 gain, suggesting alternative mechanisms of activation. Similar observation has been previously reported in breast cancers (Arnold & Papanikolaou 2005), keratoacanthoma (Burnworth et al. 2006), and squamous cell carcinoma of the skin (Utikal et al. 2005). It has also been proposed that in most cancer types, pathogenic activation of cyclin D1 can occur via additional mechanisms, including transcriptional and post-transcriptional dysregulation by oncogenic signals (Arnold & Papanikolaou 2005). Consistent with this possibility, in vitro experiments have shown direct or indirect activation of the CCND1 promoter or cyclin D1 expression by several molecules such as b-catenin, c-Jun, PPARg, calveolin-1, Ras signaling, and others (Arnold & Papanikolaou 2005). Intriguingly, three putative microRNAs (miR-1, miR-206, and miR-613) were predicted to target the 3 0 UTR of CCND1 (TargetScan 4.0), pointing to additional mechanisms for regulation of cyclin D1 expression. While cyclin D1 has been shown to promote cell proliferation and drive tumorigenesis in several human cancer models (Ewen & Lamb 2004), little is known about its role in thyroid cancer. In this study, introduction of cyclin D1 to normal thyroid cells (Nthy-ori 3-1) resulted in an increased cell population when compared with control cells. However, the difference in growth rate did not reach statistical significance. Unexpectedly, the population in HTh 7 cells transfected against CCND1 siRNA was only marginally reduced when compared with cells without CCND1 knockdown. This may be attributed to the 812

swift restoration of CCND1 within 24 h of transfection against CCND1 siRNA. The rate of transcription and translation of CCND1 within the cells of both in vitro systems could vary, explaining the discrepancies between cyclin D1 protein and CCND1 mRNA expressions. The results from siRNA and overexpression of CCND1 suggest that cyclin D1 can stimulate thyroid cell proliferation, but is in itself neither a sufficient nor a necessary factor. The identification of 20q11.2 and 20q13.12 amplicons in this study corroborated our earlier findings in ATC lines (Lee et al. 2007), as well as those reported in ATC primary tumors (Wreesmann et al. 2002, Rodrigues et al. 2004) and confirmed that the amplicons of 20q are frequent events in ATC. Interestingly, the only patient who was relapse free in this study (case 12) did not exhibit 20q gain by arrayCGH. These findings suggest that 20q gain has a role in the dedifferentiation of thyroid tumors. Recently, overexpression of UBCH10, which resides in chromosomal region 20q13.12 and belongs to the E2 gene family, was shown to be involved in thyroid cell proliferation and was therefore suggested as a candidate marker and possible therapy target for ATC (Pallante et al. 2005). We observed locus gain of UBCH10 in 25% of ATC tumors by FISH analysis, which concurs with our previous observations in ATC lines (Lee et al. 2007). Gain in 20q11.2–q21 and 20q13.12–q13.31 are also characteristics of other human cancers (Hodgson et al. 2003, Weiss et al. 2003, Lockwood et al. 2007). Gain of 20q13.12 and 11q13 was recurrently found in the same ATC cases. This could result from an unbalanced translocation followed by an amplification event as observed in lymphomas (Zhu et al. 2002), or reflect tumor evolution with selection of clones amplifying growthpromoting genes in different locations as reported in breast cancers (Al-Kuraya et al. 2004). Translocations in ATC have so far only been reported in ATC cell lines, in particular involving chromosomes 11 (Lee et al. 2007). CCND1 at the 11q13 locus is also known to be frequently co-amplified with several other genes at other chromosomes in breast carcinoma and head neck and oral squamous cell cancers (Schuuring 1995). Furthermore, co-amplification of CCND1 with genes within the 11q13 cluster in oral squamous cancer has been reported (Hsu et al. 2006). A homozygous loss in the CDKN2 locus encoding CDKN2A (p16 INK4A) on chromosome 9p21 was observed in one ATC, which was associated with lack of p16 protein expression. This prompted us to investigate p16 expression in the entire panel, which revealed lack of p16 expression in 89% of the cases. www.endocrinology-journals.org

Endocrine-Related Cancer (2008) 15 801–815 The lack of copy number loss at 9p21.3 in these ATCs suggests other mechanisms for the inactivation of p16 such as methylation (Schagdarsurengin et al. 2006). The normal thyroid tissues examined in our study did not express p16. This observation is concurrent with Ball et al. (2007) where the vast majority of normal thyroid samples lacked p16 immunostaining. However, we and others observed p16 expression in welldifferentiated thyroid tumors but not in ATC (Fig. 4C; Ferru et al. 2006, Ball et al. 2007). Taken together, these findings suggest that p16 is induced in differentiated thyroid cancer and suppressed during progression toward the undifferentiated phenotype. Gain of 16p11.2 was frequently observed in this study although this region encompasses no obvious candidate oncogenes. However, this region was identified as one of the most extensively duplicated regions on chromosome 16 based on chromosome 16 genome sequencing (Martin et al. 2004). Loss of chromosomal region 13q21 was exclusive to PTCassociated ATCs (Table 3). Notably, recurrent loss of 13q21 has been reported in PTC (Kjellman et al. 2001, Wreesmann et al. 2004). Conversely, loss of 4q determined in our study has thus far only been observed in ATC (Rodrigues et al. 2004, Lee et al. 2007). Restricted loss of 5q13.2 was noted in two ATCs with low involvement of other recurrent alterations. Of note, similar finding was previously observed in an ATC line exhibiting concomitant translocation of the 5q13 region (Lee et al. 2007). Our results are consistent with previous works (Wilkens et al. 2000, Miura et al. 2003, Pallante et al. 2005, Lee et al. 2007) showing marked DNA copy number alterations and frequent gains in ATCs; suggesting high level of chromosomal instability in ATC. Previous studies have shown that well-differentiated tumors harbor fewer alterations (Hemmer et al. 1999, Wreesmann et al. 2002). Three ATC tumors harbored the common BRAF mutation V600E that is frequently observed in PTC (Kondo et al. 2006). The array-CGH profiling and BRAF mutation findings, together with previous works, further support the hypothesis previously suggested by Galera-Davidson et al. (1987), that a subset of ATCs may be derived from dedifferentiation of PTCs. Taken together, DNA copy number changes were found to be abundant in ATCs. Gains involving 20q (20q11.2 and 20q13.12) and 11q13 represent recurrent findings potentially targeting the candidate genes CCND1/cyclin D1 and UBCH10. Lack of p16 expression and overexpression of cyclin D1 are characteristics of ATCs, and cyclin D1 has a limited effect on thyroid cell proliferation. The study revealed www.endocrinology-journals.org

several recurrent copy number alterations as well as several candidate locations for tumor suppressor genes and oncogenes that are potentially involved in molecular pathogenesis of ATC.

Declaration of interest The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.

Funding This study was supported by Swedish Cancer Society, Swedish Research Council, Go¨ran Gustafsson Foundation for Research in Natural Sciences and Medicine, Gustav V Jubilee Foundation, Cancer Society in Stockholm, Stockholm County Council, and Swedish Medical Association. The SCIBLU Genomics center is supported by grants from the K & A Wallenberg Foundation and the Lund University.

Acknowledgements We thank Prof. Lars Grimelius for the diagnostic work with the thyroid tumors. We thank Dr Nils-Erik Heldin for providing HTh7, HTh 73, C 643, SW 1736, HTh 104, HTh 83, and HTh 112, and Dr Kenneth Ain for providing KAT-4. ˚ nfalk for excellent assistance in We also thank Ms Lisa A collection of tumor samples.

References Alao JP 2007 The regulation of cyclin D1 degradation: roles in cancer development and the potential for therapeutic invention. Molecular Cancer 6 24. Albertson DG 2006 Gene amplification in cancer. Trends in Genetics 22 447–455. Al-Kuraya K, Schraml P, Torhorst J, Tapia C, Zaharieva B, Novotny H, Spichtin H, Maurer R, Mirlacher M, Ko¨chi O et al. 2004 Prognostic relevance of gene amplifications and coamplifications in breast cancer. Cancer Research 64 8534–8540. Arnold A & Papanikolaou A 2005 Cyclin D1 in breast cancer pathogenesis. Journal of Clinical Oncology 23 4215–4224. Autio R, Hautaniemi S, Kauraniemi P, Yli-Harja O, Astola J, Wolf M & Kallioniemi A 2003 CGH-Plotter: MATLAB toolbox for CGH-data analysis. Bioinformatics 19 1714–1715. Ball E, Bond J, Franc B, Demicco C & Wynford-Thomas D 2007 An immunohistochemical study of p16(INK4a) expression in multistep thyroid tumourigenesis. European Journal of Cancer 43 194–201. Barbaro M, Oscarson M, Schoumans J, Staaf J, Ivarsson SA & Wedell A 2007 Isolated 46,XY gonadal dysgenesis in

813

J-J Lee et al.: Genetic alterations in anaplastic thyroid cancer two sisters caused by a Xp21.2 interstitial duplication containing the DAX1 gene. Journal of Clinical Endocrinology and Metabolism 92 3305–3313. Burnworth B, Popp S, Stark HJ, Steinkraus V, Brocker EB, Hartschuh W, Birek C & Boukamp P 2006 Gain of 11q/cyclin D1 overexpression is an essential early step in skin cancer development and causes abnormal tissue organization and differentiation. Oncogene 25 4399–4412. DeLellis RA, Lloyd RV, Heitz PU & Eng C 2004 Pathology and Genetics of Tumours of Endocrine Organs. Lyon: IARC Press. Ewen ME & Lamb J 2004 The activities of cyclin D1 that drive tumorigenesis. Trends in Molecular Medicine 10 158–162. Ferru A, Fromont G, Gibelin H, Guilhot J, Savagner F, Tourani JM, Kraimps JL, Larsen CJ & Karayan-Tapon L 2006 The status of CDKN2A alpha (p16INK4A) and beta (p14ARF) transcripts in thyroid tumour progression. British Journal of Cancer 95 1670–1677. Galera-Davidson H, Bibbo M, Dytch HE, Gonzalez-Campora R, Fernandez A & Wied GL 1987 Nuclear DNA in anaplastic thyroid carcinoma with a differentiated component. Histopathology 11 715–722. Garcia-Rosta´n G, Costa AM, Pereira-Castro I, Salvatore G, Hernandez R, Hermsem MJ, Herrero A, Fusco A, Cameselle-Teijeiro J & Santoro M 2005 Mutation of the PIK3CA gene in anaplastic thyroid cancer. Cancer Research 65 10199–10207. Hemmer S, Wasenius VM, Knuutila S, Franssila K & Joensuu H 1999 DNA copy number changes in thyroid carcinoma. American Journal of Pathology 154 1539–1547. Hodgson JG, Chin K, Collins C & Gray JW 2003 Genome amplification of chromosome 20 in breast cancer. Breast Cancer Research and Treatment 78 337–345. Hou P, Liu D, Shan Y, Hu S, Studeman K, Condouris S, Wang Y, Trink A, El-Naggar AK, Tallini G et al. 2007 Genetic alterations and their relationship in the phosphatidylinositol 3-kinase/Akt pathway in thyroid cancer. Clinical Cancer Research 13 1161–1170. Hsu L-C, Huang X, Seasholtz S, Potter DM & Gollin SM 2006 Gene amplification and overexpression of protein phophatase 1a in oral squamous cell carcinoma cell lines. Oncogene 25 5517–5526. Jo¨nsson G, Dahl C, Staaf J, Sandberg T, Bendahl PO, Ringne´r M, Ho¨glund M & Borg A 2007 Genomic profiling of malignant melanoma using tiling-resolution arrayCGH. Oncogene 26 4738–4748. Khoo ML, Ezzat S, Freeman JL & Asa SL 2002 Cyclin D1 protein expression predicts metastatic behavior in thyroid papillary microcarcinomas but is not associated with gene amplification. Journal of Clinical Endocrinology and Metabolism 87 1810–1813. Kjellman P, Lagercrantz S, Hoog A, Wallin G, Larsson C & Zedenius J 2001 Gain of 1q and loss of 9q21.3–q32 are associated with a less favorable prognosis in papillary thyroid carcinoma. Genes, Chromosomes and Cancer 32 43–49.

814

Kondo T, Ezzat S & Asa SL 2006 Pathogenetic mechanisms in thyroid follicular-cell neoplasia. Nature Reviews. Cancer 6 292–306. Lee JJ, Foukakis T, Hashemi J, Grimelius L, Heldin NE, Wallin G, Rudduck C, Lui WO, Ho¨o¨g A & Larsson C 2007 Molecular cytogenetic profiles of novel and established human anaplastic thyroid carcinoma models. Thyroid 17 289–301. Lockwood WW, Coe BP, Williams AC, MacAulay C & Lam WL 2007 Whole genome tiling path array CGH analysis of segmental copy number alterations in cervical cancer cell lines. International Journal of Cancer 120 436–443. Martin J, Han C, Gordon LA, Terry A, Prabhakar S, She X, Xie G, Hellsten U, Chan YM, Altherr M et al. 2004 The sequence and analysis of duplication-rich human chromosome 16. Nature 432 988–994. Miura D, Wada N, Chin K, Magrane GG, Wong M, Duh QY & Clark OH 2003 Anaplastic thyroid cancer: cytogenetic patterns by comparative genomic hybridization. Thyroid 13 283–290. Pallante P, Berlingieri MT, Troncone G, Kruhoffer M, Orntoft TF, Viglietto G, Caleo A, Migliaccio I, Decaussin-Petrucci M, Santoro M et al. 2005 UbcH10 overexpression may represent a marker of anaplastic thyroid carcinomas. British Journal of Cancer 93 464–471. Rodrigues RF, Roque L, Rosa-Santos J, Cid O & Soares J 2004 Chromosomal imbalances associated with anaplastic transformation of follicular thyroid carcinomas. British Journal of Cancer 90 492–496. Saal LH, Troein C, Vallon-Christersson J, Gruvberger S, Borg A & Peterson C 2002 BioArray Software Environment (BASE): a platform for comprehensive management and analysis of microarray data. Genome Biology 3 SOFTWARE0003. Schagdarsurengin U, Gimm O, Dralle H, Hoang-Vu C & Dammann R 2006 CpG island methylation of tumorrelated promoters occurs preferentially in undifferentiated carcinoma. Thyroid 16 633–642. Schuuring E 1995 The involvement of the chromosome 11q13 region in human malignancies: cyclin D1 and EMS1 are two new candidate oncogenes – a review. Gene 159 83–96. Utikal J, Udart M, Leiter U, Peter RU & Krahn G 2005 Additional Cyclin D(1) gene copies associated with chromosome 11 aberrations in cutaneous malignant melanoma. International Journal of Oncology 26 597–605. Wallin G, Lundell G & Tennvall J 2004 Anaplastic giant cell thyroid carcinoma. Scandinavian Journal of Surgery 93 272–277. Wang S, Lloyd RV, Hutzler MJ, Safran MS, Patwardhan NA & Khan A 2000 The role of cell cycle regulatory protein, cyclin D1, in the progression of thyroid cancer. Modern Pathology 13 882–887. Weiss MM, Snijders AM, Kuipers EJ, Ylstra B, Pinkel D, Meuwissen SG, van Diest PJ, Albertson DG & Meijer GA 2003 Determination of amplicon boundaries at 20q13.2 in

www.endocrinology-journals.org

Endocrine-Related Cancer (2008) 15 801–815 tissue samples of human gastric adenocarcinomas by high-resolution microarray comparative genomic hybridization. Journal of Pathology 200 320–326. Wilkens L, Benten D, Tchinda J, Brabant G, Potter E, Dralle H & von Wasielewski R 2000 Aberrations of chromosomes 5 and 8 as recurrent cytogenetic events in anaplastic carcinoma of the thyroid as detected by fluorescence in situ hybridisation and comparative genomic hybridization. Virchows Archiv 436 312–318. Wreesmann VB, Ghossein RA, Patel SG, Harris CP, Schnaser EA, Shaha AR, Tuttle RM, Shah JP, Rao PH &

www.endocrinology-journals.org

Singh B 2002 Genome-wide appraisal of thyroid cancer progression. American Journal of Pathology 161 1549–1556. Wreesmann VB, Sieczka EM, Socci ND, Hezel M, Belbin TJ, Childs G, Patel SG, Patel KN, Tallini G, Prystowsky M et al. 2004 Genome-wide profiling of papillary thyroid cancer identifies MUC1 as an independent prognostic marker. Cancer Research 64 3780–3789. Zhu C, Mills KD, Ferguson DO, Lee C, Manis J, Fleming J, Gao Y, Morton CC & Alt FW 2002 Unrepaired DNA breaks in p53-deficient cells lead to oncogenic gene amplification subsequent to translocations. Cell 109 811–821.

815